author query form - lsbu open research...e-mail: corrections.esch@elsevier.thomsondigital.com...
Post on 29-Nov-2020
2 Views
Preview:
TRANSCRIPT
Our reference: ENB 4676 P-authorquery-v9
AUTHOR QUERY FORM
Journal: ENB Please e-mail or fax your responses and any corrections to:
E-mail: corrections.esch@elsevier.thomsondigital.com
Article Number: 4676 Fax: +353 6170 9272
Dear Author,
Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screenannotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other thanAdobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper pleasereturn your corrections within 48 hours.
For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions.
Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags inthe proof. Click on the ‘Q’ link to go to the location in the proof.
Location in Query / Remark: click on the Q link to goarticle Please insert your reply or correction at the corresponding line in the proof
Reference(s) given here were noted in the reference list but are missing from the text – please positioneach reference in the text or delete it from the list.
Q1 Please confirm that given names and surnames have been identified correctly.Q2 Uncited reference: This section comprises reference that occur in the reference list but not in the body
of the text. Please position the reference in the text or, alternatively, delete it. Any reference not dealtwith will be retained in this section.
Q3 References provided in the appendix section has been moved to reference section. Kindly check forcorrectness.
Please check this box or indicate your approval ifyou have no corrections to make to the PDF file
Thank you for your assistance.
ENB 4676 1
ARTICLE IN PRESSG Model
Energy and Buildings xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Energy and Buildings
j o ur nal ho me page: www.elsev ier .com/ locate /enbui ld
Graphical Abstract
Energy and Buildings xxx (2013) xxx–xxxSpecific energy consumption values for various refrigerated food cold stores
J.A. Evans∗, A.M. Foster, J.-M. Huet, L. Reinholdt, K. Fikiin, C. Zilio, M. Houska,A. Landfeld, C. Bond, M. Scheurs, T.W.M. van Sambeeck
ENB 4676 1
ARTICLE IN PRESSG Model
Energy and Buildings xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Energy and Buildings
j o ur nal ho me page: www.elsev ier .com/ locate /enbui ld
Highlights
Energy and Buildings xxx (2013) xxx–xxxSpecific energy consumption values for various refrigerated food cold stores
J.A. Evans∗, A.M. Foster, J.-M. Huet, L. Reinholdt, K. Fikiin, C. Zilio, M. Houska, A. Landfeld, C. Bond, M. Scheurs, T.W.M. van Sambeeck
• Energy consumption of cold stores was compared and benchmarked.• The work consists of the greatest number of data sets collected and published to date.• A strong relationship between volume and energy was established.• A mathematical model was developed to predict energy use.• The model was used to identify factors affecting energy consumption.
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
Energy and Buildings xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Energy and Buildings
j ourna l ho me p age: www.elsev ier .com/ locate /enbui ld
Specific energy consumption values for various refrigeratedfood cold stores
1
2
J.A. Evansa,∗, A.M. Fostera, J.-M. Huetb,1, L. Reinholdtb,1, K. Fikiinc,2, C. Ziliod,3,Q1
M. Houskae,4, A. Landfelde,4, C. Bondf,5, M. Scheursg,6, T.W.M. van Sambeeckh,73
4
a Faculty of Engineering, Science and the Built Environment, London South Bank University, Langford, Bristol BS40 5DU, UK5b Danish Teknologisk Institut, Kongsvang Allé 29, 8000 Aarhus C, Denmark6c Technical University of Sofia, 1000, 8 Kliment Ohridsky Str., Sofia, Bulgaria7d University of Padova, Department of Industrial Engineering, Via Venezia, 1, I-35131 Padova, Italy8e Food Research Institute Prague, Radiova str. 7, 102 31 Prague 10, Czech Republic9f Carbon Data Resources Ltd., Carbon House, 15 The Batch, Bath, Batheaston, Somerset BA1 7DR, UK10g Catholic University College Limburg, Campus Diepenbeek, Agoralaan Gebouw B, 3590 Diepenbeek, Belgium11h Cold Chain Experts in ColdstoreExpertiseCenter, Van Dongenshoeve 10, 8052 BJ Hattem, The Netherlands12
13
a r t i c l e i n f o14
15
Article history:16
Received 15 February 201317
Received in revised form18 November 2013
18
19
Accepted 22 November 201320
21
Keywords:22
Refrigeration23
Cold store24
Food25
Energy26
Specific energy consumption27
Mathematical model28
a b s t r a c t
Two benchmarking surveys were created to collect data on the performance of chilled, frozen and mixed(chilled and frozen stores operated from a single refrigeration system) food cold stores with the aim ofidentifying the major factors influencing energy consumption. The volume of the cold store was foundto have the greatest relationship with energy use with none of the other factors collected having anysignificant impact on energy use. For chilled cold stores, 93% of the variation in energy was related tostore volume. For frozen stores, 56% and for mixed stores, 67% of the variation in energy consumptionwas related to store volume. The results also demonstrated the large variability in performance of coldstores. This was investigated using a mathematical model to predict energy use under typical cold storeconstruction, usage and efficiency scenarios. The model demonstrated that store shape factor (which hada major impact on surface area of the stores), usage and to a lesser degree ambient temperature all hadan impact on energy consumption. The work provides an initial basis to compare energy performanceof cold stores and indicates the areas where considerable energy saving are achievable in food coldstores.
© 2013 Published by Elsevier B.V.
29
1. Introduction30
Refrigeration is one of the most energy-intensive technolo-31
gies used in the food supply chain and poses a number of32
sustainability-related challenges. It accounts for about 35% of33
∗ Corresponding author. Tel.: +44 117928 9300.E-mail addresses: j.a.evans@lsbu.ac.uk (J.A. Evans), a.m.foster@lsbu.ac.uk
(A.M. Foster), JHU@teknologisk.dk (J.-M. Huet), lre@teknologisk.dk(L. Reinholdt), agf@tu-sofia.bg (K. Fikiin), claudio.zilio@unipd.it (C. Zilio),milan.houska@vupp.cz (M. Houska), ales.landfeld@vupp.cz (A. Landfeld),carole@carbon-data.co.uk (C. Bond), marc.schreurs@khlim.be (M. Scheurs),tvansambeeck@ColdstoreExpertiseCenter.com (T.W.M. van Sambeeck).
1 Tel.: +45 72201270.2 Tel.: +359 895586030.3 Tel.: +39 0498276893.4 Tel.: +420 296792306.5 Tel.: +44 1225852533.6 Tel.: +32 11230770.7 Tel.: +31 384446186.
electricity consumption in the food industry [1], worldwide this 34
equates to a consumption of about 1300 TWh year−1 [1]. 35
Energy issues are among the main concerns in Europe today. 36
The main challenge is to meet the binding target set by the Heads 37
of States and Governments of the 27 EU Member States in March 38
2007 to increase energy efficiency by 20% and to increase the use 39
of renewable energies by 20%, by 2020 [2]. 40
All chilled and frozen food and temperature controlled phar- 41
maceutical products are stored in a cold store at least once during 42
their journey from production to the consumer. Chilled stores gen- 43
erally maintain products at temperatures between −1 and 10 C 44
whereas frozen stores generally maintain product at below −18 C. 45
The cold store market is extremely diverse consisting of small stores 46
of 10–20 m3 up to large warehouses of hundreds of thousands of 47
cubic metres. All cold stores have the function of storing a product 48
at the correct temperature and to prevent quality loss as economi- 49
cally as possible. In Europe there are approximately 1.7 million cold 50
stores totalling 60–70 million m3 of storage volume. Of these, 67% 51
are small stores with a volume of less than 400 m3 [3]. 52
0378-7788/$ – see front matter © 2013 Published by Elsevier B.V.http://dx.doi.org/10.1016/j.enbuild.2013.11.075
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
2 J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx
Cold storage rooms consume considerable amounts of energy.53
Previous unpublished work by the authors has shown that within54
cold storage facilities, 60–70% of the electrical energy may be55
used for refrigeration. Therefore, cold store users have consid-56
erable incentive to reduce energy consumption. There are few57
published surveys comparing the performance of more than a few58
cold stores. In addition surveys rarely differentiate between type59
of store, storage temperature, location, room size or room function.60
In 2002 the IIR estimated that the SEC (Specific Energy Consump-61
tion) of cold stores was between 30 and 50 kWh m−3 year−1 [4].62
The minimum value from this study was similar to values from a63
study carried out in the Netherlands by Bosma [5] which found64
energy consumption of cold stores to be 35 kWh m−3 year−1. In65
the UK ETSU (Energy Technology Savings Unit) [6] also found that66
stores consumed at minimum 34 kWh m−3 year−1 but that con-67
sumption could also be up to 124 kWh m−3 year−1. Other studies68
in the USA by Elleson and Freund [7] and Singh [8] found SECs69
of between 19 and 88, and 15 and 132 kWh m−3 year−1 respec-70
tively. In one of the most comprehensive recent surveys carried71
out in New Zealand by Werner et al. [9] the performance of 3472
cold stores was compared. The SECs recorded varied from 26 to73
379 kWh m−3 year−1 demonstrating that there was a large varia-74
tion in energy consumed by cold stores. Savings of between 1575
and 26% were found to be achievable by applying best practice76
technologies. This large range in performance was also found by77
Carlsson-Kanyama and Faist [10] who report data from BELF [11]78
for energy use for freezers per litre net volume per day to be79
1.0 kJ (equivalent to 101 kWh m−3 year−1) when food was stored80
in rooms of 10,000 m3 whereas in rooms of 10 m3 the energy was81
15 kJ (equivalent to 1520 kWh m−3 year−1). In both surveys a factor82
difference of 15 was apparent.83
Limited information has been published on throughputs and84
storage and often information is difficult to compare due to the met-85
rics used by the authors. Carlsson-Kanyama and Faist [10] report86
energy used for long-term cold storage of apples may vary between87
0.9–1.7 kJ electricity per kg per day. Swian [12] reported figures88
for potato storage collected over a 3 year period from 8 stores as89
being between 0.1 and 0.29 kWh tonne−1 day−1. On average the90
energy ranged from 0.12 to 0.15 kWh tonne−1 day−1 within each91
of the 3 years where monitoring took place. The results showed92
a massive difference in energy consumption between the best93
and worse stores. It should be noted that the data included all94
energy used and that in cold weather potato farmers need to heat95
stores to maintain the potatoes at the usual storage tempera-96
tures of 3 C. In addition there was no information presented on97
store temperatures and so the stores that appear most efficient98
may be those that stored the potatoes at a higher tempera-99
ture.100
Previous detailed audits carried out on a small number of cold101
stores has confirmed that energy consumption can vary consider-102
ably and that this was due to a variety of factors [13,14]. These103
surveys also demonstrated that energy savings of 30–40% were104
achievable by optimising usage of the stores, repairing current105
equipment and by retrofitting of energy efficient equipment.106
The performance of a large number of cold stores has never been107
compared in detail and there is little information to compare per-108
formance of stores Worldwide. With government targets to reduce109
energy and emissions of greenhouse gasses (GHG), the need to110
benchmark and understand potential energy and GHG reductions111
is of great interest to end users. To enable end users to improve the112
performance of their cold stores a project called ‘Improving Cold113
storage Equipment in Europe’ (ICE-E) was developed with 8 part-114
ners from across Europe. The initial aim of the project was to collect115
data to benchmark the performance of cold stores in Europe.116
As part of the ICE-E project, two internet based surveys117
were developed and data collected to determine energy usage in118
different cold store types, sizes and configurations. In addition a 119
mathematical model was developed to predict energy used in cold 120
stores. Results from these surveys and the predictions made by 121
the model are presented in this paper and the data analysed to 122
determine whether there were any common factors that affected 123
performance of the cold stores. 124
2. Materials and methods 125
2.1. Detailed survey tool 126
2.1.1. Development of survey tool 127
The survey was developed using a NET web application. Devel- 128
opment was carried out in Microsoft Visual Studio using c# (c sharp) 129
which used .NET Framework 4.0. The data was saved in a Microsoft 130
SQL database. The survey was available in a number of languages 131
(Bulgarian, Czech, Danish, Dutch, English, French, Italian and Span- 132
ish). The survey was initially tested on a selected number of cold 133
store operators to ensure the questions were appropriate and rel- 134
evant. Improvements were then made based on their comments. 135
The survey allowed participants to initially register their details 136
and then to enter data on as many refrigeration systems as they 137
wished. It collected information per single refrigeration system that 138
might supply one of several cold stores. The survey was designed 139
to be simple to complete with the aim that it should take a cold 140
store operator less than 20 min to complete the survey. The final 141
survey document consisted of 5 pages collecting basic information, 142
information on the refrigeration system, the food stored, the facility 143
and the refrigeration equipment at the facility. During the initial 144
registration process, cold store operators could ensure that data 145
was anonymous. 146
2.1.2. Data collected and benchmark analysis of survey tool 147
The survey parameters collected are shown in Appendix 1. In 148
all cases the users were asked to rate the accuracy of the data they 149
submitted. The collected data was retained on a server where users 150
could return to update information or add further data. 151
Once users had input data they could then compare the per- 152
formance of their store through an automatic benchmark analysis. 153
This enabled them to compare the energy used by their cold store 154
system with systems of a similar size and product throughput. In 155
addition users could compare the set point temperatures, food type, 156
room function and refrigerant type with others in the survey. In all 157
comparisons the user had the ability to define the range over which 158
comparisons were carried out. 159
2.2. Express survey tool 160
In response to some end users requesting a simpler and more 161
rapid means to benchmark their stores an ‘Express Survey’ was 162
developed. This required only 5 min to complete. 163
2.2.1. Development of survey tool 164
The tool was part of the ICE-E web site and written in HyperText 165
Markup Language (HTML) using a web form to collect the data. As 166
in the detailed survey all data collected was anonymous. 167
2.2.2. Data collected and benchmark analysis of survey tool 168
A limited data set of 5 parameters was collected (set temper- 169
ature, area and volume of the store, food throughput and energy 170
usage per year) which reflected what were considered to be the 171
most important factors affecting energy use in cold stores. In all 172
cases blast freezing of product was excluded from the data col- 173
lected. 174
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx 3
Once data was submitted the information was input manually175
into the main benchmark survey and information sent directly to176
the cold store operator.177
For both surveys the data collected was checked and unreli-178
able data excluded. Where possible any unreliable data was cross179
checked with the cold store operator and any anomalies corrected.180
2.2.3. Mathematical model of cold store energy performance181
A mathematical model of cold store energy use was developed182
to predict energy used by cold stores. This was used to compare183
theoretical energy used by cold stores with the actual energy usage184
collected in the survey.185
The model was steady state, therefore all heat loads were aver-186
aged over one day. The cold store was modelled as a fully sealed187
rectangular box with one entry door. The cold store had enough188
thermal mass such that door openings did not change the tem-189
perature in the cold store. The temperature of the ambient air190
outside the cold store was not changed by the door openings. There191
was only one layer of insulation on the walls, roof and floor. Any192
metal cladding was ignored as the resistance to heat transfer from193
this was considered negligible. The luminous flux from the lights194
was divided by the area of the floor and walls to give a uniform195
luminance. The thermal mass of the forklift trucks were ignored.196
Therefore if they moved from a warm environment into the store,197
they did not give up this heat to the store. Energy from fork lift198
trucks did not include charging the batteries. Any product which199
changed temperature when loaded into the store did not have a200
latent load (e.g. freezing and thawing), only a sensible load.201
Data was input via a spread sheet. The inputs included;202
• Information about each wall (including ceiling and floor) of the203
cold store, e.g. face area, whether it was in the sun, outside ambi-204
ent or internal and the type and thickness of the insulation.205
• The size of the door, its opening schedule, whether it was pro-206
tected (e.g. by strip or curtains), amount of traffic through the207
door and the outside conditions.208
• The refrigeration system, refrigerant, type of condenser, con-209
denser ambient, efficiency of compressor and number of210
compression stages.211
• Heat loads inside the store from forklifts, lights, personnel, prod-212
uct, defrosts and evaporator fans.213
• Electrical loads from lights, defrosts, evaporator fans and con-214
denser fans.215
Full details of the model are contained in Appendix 2.216
To better understand the variations in the survey data, 3 usage217
scenarios were modelled over a range of store volumes between 10218
and 350,000 m3. Store volume was modelled as a cold store of 5 m219
height with store width and depth equal in all cases. A further set220
of predictions were made at each store volume for the stores with221
the minimum and maximum practical surface area (an assump-222
tion was made that the store height could not be less than 2 m). For223
each scenario a chilled store at 2 C and a frozen store at −23 C were224
modelled at the minimum and maximum average annual tempera-225
tures in Europe (4.6 C and 20.6 C based on data from weatherbase226
[16]). The 3 scenarios were:227
1. A base-line store where all heat loads except those that were228
essential to the operation of the store were removed.229
2. A typical store with average use with a high efficiency refriger-230
ation system.231
3. A typical store with high usage with a low efficiency refrigeration232
system.233
Parameters for each scenario were selected based on informa- 234
tion from Evans et al. [17]. Full details of the assumptions made for 235
each of the 3 scenarios are listed in Table 1. 236
3. Results 237
3.1. Data collected 238
Data from 329 cold stores was collected. One data point was 239
the mean of 331 cold stores in the UK (i.e. the total data collection 240
encompassed 659 stores). This point was excluded from the anal- 241
ysis as data was not available on the data variance. Therefore, the 242
data point could not be included at an equal weighting to the other 243
data sets and so was used for purely comparative purposes in the 244
analysis. Thirty-four data sets were removed as they were consid- 245
ered unreliable (due to store dimensions being obviously incorrect 246
or product temperatures, throughputs or store temperatures being 247
inconsistent) leaving 294 data sets with the minimum 5 critical 248
parameters recorded (temperature of the store, area and volume of 249
the store, food throughput and energy usage per year). 250
The data collected covered 21 different countries (Belgium, Bul- 251
garia, China, Czech Republic, Denmark, France, Germany, Greece, 252
Ireland, Italy, Mexico, Netherlands, New Zealand, Portugal, Roma- 253
nia, Serbia, Spain, Sweden, Switzerland, United Kingdom, USA). 254
Seventy percent of the 295 data sets originated from EU countries. 255
3.2. Cold store type 256
Cold store function was divided into chilled, frozen or mixed 257
stores (those with both chilled and frozen rooms operating from 258
a common refrigeration system). Analysis of variance (ANOVA) 259
showed a highly significant difference (P < 0.05) between the SEC of 260
all store types. Differences between chilled and frozen and chilled 261
and mixed were greater (P < 0.01) than between frozen and mixed 262
stores (P < 0.05). 263
3.3. Country 264
Large variations in SEC were shown between countries. How- 265
ever, this was most likely due to the limited number of data sets for 266
some countries. Analysing the data from countries, where a greater 267
number of data sets were available, did not show any correlation 268
between location and ambient temperature at the location or any 269
factor such as differences in design of the cold stores. Due to the 270
large variability in SEC it was not possible to analyse data from 271
each country separately. Therefore, all further analysis was carried 272
out on data divided into chilled, frozen and mixed stores. 273
3.4. Impact of store location and ambient temperature 274
An analysis of ambient temperature at each store location 275
was carried out. Data on ambient temperature was taken from 276
meteorological data for each store location and the mean annual 277
temperature for the year in which the energy data was collected 278
was correlated with energy usage. Correlations between ambient 279
temperature and SEC for chilled, frozen and mixed stores were low 280
(less than 0.17), indicating that mean ambient temperature may 281
have had little impact on energy usage. 282
3.5. Relationship between energy use and store size 283
The relationship between store energy consumption and size 284
was investigated using multiple regression. As part of this analysis 285
the data was found to be near to a normal distribution. 286
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
4 J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx
Table 1Assumptions used in model.
Scenario 1 Scenario 2 Scenario 3
Chilled Frozen Chilled Frozen Chilled Frozen
Cold store shading Shaded Not shaded Not shadedCold store colour Light Dark DarkInsulation 100 mm 150 mm 100 mm 150 mm 100 mm 150 mmAir movement around store Still air Windy WindyUnder floor heating None None NoneRefrigerant R717 (ammonia) R717 (ammonia) R717 (ammonia)Condenser Evaporative Evaporative AirCompression/expansion stages 2 Compressor/1
expansion stage2 Compressor/1expansion stage
2 Compressor/1expansion stage
Isentropic efficiencyof compressor
High (0.7) High (0.7) Low (0.5)
Defrost Off-cycle Electric Off-cycle Electric Off-cycle ElectricProduct heat load None Food loaded at 1 C above store
temperature, loading = 250 kg m−3,product weight loss = zero
Food loaded at 2 C above storetemperature, loading = 500 kg m−3,product weight loss = zero
25% of total massloaded each day
10% of total massloaded each day
25% of total massloaded each day
10% of total massloaded each day
Fork lift heat load None 1 per 40,000 m3, size = medium,electric, operated 24 h per day
1 per 30,000 m3, size = medium,electric, operated 24 h per day
People heat load None 2 persons per forklift truck, 24 hoursper day
2 persons per forklift truck, 24 hoursper day
Lighting heat load None Fluorescent lights, 50 lumens.W−1, 500lux, operational 24 hours per day
Fluorescent lights, 50 lumens.W−1, 500lux, operational 24 hours per day
Infiltration heat load None Door height 2.5 m, width 2 mminimum, if > 50,000 m3 store volumethen door width = store volume/10,000,door opening time = 25 sec, volume oftraffic during door opening = medium,door seal = good, no protection on door
Door height 2.5 m, width 2 mminimum, if > 50,000 m3 store volumedoor width = store volume/10,000,door opening time = 25 sec, volume oftraffic during door opening = medium,door seal = good, no protection on door
400 door openings perday
200 door openings perday
600 door openings perday
300 door openings perday
Evaporator/condenserfan power
Created from correlation from Evanset al. [17]
Same as for scenario 1 Same as for scenario 1
3.5.1. Chilled stores287
One hundred and twenty-six chilled stores were included in the288
analysis. These ranged in volume from 57 to 225,000 m3. Regression289
demonstrated that 93% of the variation in annual energy con-290
sumption was related to store volume (Fig. 1). Multiple regression291
demonstrated that food type and food throughput had some impact292
on annual energy but that these factors only increased the R2 value293
to 95% and therefore their impact was very low. All other factors294
collected (including store temperature, store insulation type and295
thickness, store location and ambient conditions around the store,296
type of refrigerant and effect of door protection) had no influence297
on annual energy consumption.298
Applying non linear relationships to the data did not improve299
the regression R2 value. This indicates that SEC remained relatively300
constant across the range of cold store volumes examined.301
3.5.2. Frozen stores302
One hundred and thirty-two frozen stores were included in the303
analysis. These ranged in volume from 100 to 291,280 m3. Store304
volume accounted for 56% of the variability in annual energy con- 305
sumption of frozen stores when a linear regression was applied. 306
Applying a non linear power function to the data improved the 307
regression R2 value to 66% (Fig. 2). This showed that for frozen stores 308
SEC reduced as the store size increased. 309
As with chilled stores none of the factors recorded had any- 310
thing above a very minimal impact on annual energy consumption. 311
Therefore, approximately 34% of the variability in annual energy 312
consumption was related to a factor that was not collected in the 313
survey. 314
3.5.3. Mixed stores 315
Thirty-six mixed stores were included in the analysis. These 316
ranged in volume from 9100 to 180,000 m3. A number of factors 317
had an impact on mixed store annual energy consumption. As a 318
linear regression, store volume accounted for 67% of the variability, 319
however, if a power function (non linear regression) was applied 320
this increased to 76%. (Fig. 3). In addition throughput, thickness of 321
the store insulation (wall, ceiling and floor) and insulation age also 322
Table 2Range in SEC values for cold stores examined.
Chilled(kWh m−3 year−1)
Frozen(kWh m−3 year−1)
Mixed(kWh m−3 year−1)
All data Mean 56.1 73.5 61.2Minimum 4.4 6.0 15.7Maximum 250.4 240.4 115.8
10% upper and 10% lower values removed Mean 52.2 66.7 52.0Minimum 20.7 27.0 29.9Maximum 95.3 134.6 107.0
20% upper and 20% lower values removed Mean 51.5 63.7 55.9Minimum 29.6 37.5 37.4Maximum 78.0 100.0 87.3
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx 5
Fig. 1. Relationship between store volume and total energy use per year (kWh/year) for chilled stores.
appeared to have a minor impact on annual energy consumption.323
However, for these data sets the number of replicates was low and324
so their impact needs further investigation.325
Mixed stores appeared to have a similar volume relationship326
with annual energy consumption as frozen stores and therefore327
the store SEC reduced for larger stores.328
3.6. All stores329
The SEC for the cold stores examined varied considerably. Data330
from all stores and for all stores with the 10% and 20% upper and331
lower values removed are shown in Table 2.332
It is interesting to note that mixed and frozen stores had a rel-333
atively similar relationship between volume and annual energy334
(although statistically the regression lines were significantly differ-335
ent at P < 0.01). At volumes below 22,000 m3 chilled store used less336
energy than frozen or mixed stores but at volumes above 22,000 m3337
chilled stores used more energy than frozen or mixed stores. The338
stores below 22,000 m3 were dominated by a cluster of smaller339
chilled stores that had low energy consumption (several of them340
were produce stores where there was often intermittent usage). It 341
would be expected that chilled stores would use less energy than 342
frozen stores across the whole range of volumes. However, it may 343
be reasonable to expect that chilled stores have greater usage and 344
greater product heat loads than frozen stores which tend to be used 345
for long term storage of food. 346
3.6.1. Mathematical model of cold store energy performance 347
Results from the 3 modelling scenarios are presented in Fig. 4 348
(chilled stores) and Fig. 5 (frozen stores). The dashed area out- 349
lined in each figure represents the range in energy consumption 350
for stores with varied shape factors predicted by the model. Shape 351
factor is related to store surface area and was found to have a large 352
impact on energy consumption and was responsible for increas- 353
ing energy use from the most efficient shape (lowest surface area 354
to volume ratio, in this case a cube) to the least (a flat plane) by 13 355
times in a chiller and 10 times in a freezer. The differences between 356
chillers and freezers were due to insulation thickness (most com- 357
monly 100 mm in a chiller and 150 mm in a freezer). 358
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
6 J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx
Fig. 2. Relationship between store volume and total energy use per year (kWh/year) for frozen stores (non linear regression).
Ambient conditions around the store had greater influence on359
frozen stores than chilled stores (due to the greater temperature360
difference between ambient and store temperature for freezers).361
Ambient temperature had a greater impact on energy use when362
usage of the store was high (due to door openings).363
If the predications made by the model (including all scenarios364
and range due to ambient temperature and shape factor) were com-365
pared to the data collected in the survey the model predictions366
covered 83% of frozen stores and 94% of chilled stores. Assuming367
some inaccuracies in the model (±10%) a further 2% of chilled stores368
and 6% of frozen stores would be included within the predicted369
ranges. The stores that were outside of this predicted range were370
stores with small volumes (less than 8270 m3 for chilled stores and371
less than 30,000 m3 for frozen stores).372
SEC decreased as store volume increased but the reduction in373
SEC was most apparent at low store volumes. The SEC changed by374
less than 0.5 kWh m−3 year−1 per 10,000 m3 increase in store vol-375
ume for stores with volumes of greater than 10,000 m3 for chillers376
and 20,000 m3 for freezers. Due the minimal change in SEC above 377
certain store volumes the relationship between energy and vol- 378
ume predicted by the model approached a linear relationship when 379
considering store sizes of up to 350,000 m3. 380
3.6.2. Use of the model to assess the efficiency of cold stores 381
Using the knowledge gained from the model, the energy used by 382
the survey cold stores could be compared to the modelled energy 383
usage. As the total store surface area was found to be a factor in 384
the energy usage, the energy used across a range of total store 385
surface areas, usage scenarios and ambient temperatures was pre- 386
dicted by the model. Total surface area was obtainable for the cold 387
stores modelled but had to be estimated for the survey data. The 388
floor and ceiling surface area was recorded in the survey (ceiling 389
area was assumed to be equal to floor area). The area of each wall 390
was estimated by multiplying store height (obtained by dividing 391
the store volume by the area) by store depth/width (calculated 392
by taking the square root of the store area). For each cold store 393
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx 7
Fig. 3. Relationship between store volume and total energy use (kWh/year) for mixed stores (non linear regression).
survey data point, the average annual ambient temperature for394
the cold store location was extracted from a weather database395
(http://www.weatherbase.com/).396
The impact of total surface area on the energy consumed for each397
modelled scenario is shown in Fig. 6 (for chilled stores) and Fig. 7398
(for frozen stores). This can then be used to estimate the energy399
that a cold store should use in a particular ambient location and400
with a particular usage. The modelled results were compared to401
the survey data for ambient temperature surrounding the store of402
12.6 ± 4 C and are presented in Figs. 6 and 7. The results show that403
even though full details of usage for the survey population are not404
known, that some stores consume considerably more and some less405
energy than is predicted. By using this methodology the divergence406
between the energy actually used by a cold store and the energy it407
should use can be identified. This could be used to provide a metric408
of energy use per year per square metre that can be used to assess409
operation of cold stores.
4. Discussion 410
The data collected showed that there was large vari- 411
ability in the energy used by cold stores. The SEC varied 412
between 4 and 250 kWh m−3 year−1 for chillers, between 6 413
and 240 kWh m−3 year−1 for freezers and between 23 and 414
157 kWh m−3 year−1 for mixed stores. The minimum SEC val- 415
ues for chilled and frozen stores were lower than have been 416
reported previously by most authors [4–9] but were not dis- 417
similar to those reported by Carlsson-Kanyama and Faist [10]. 418
However, the maximum SEC values were greater than reported 419
by the IIR [4], Bosma [5], ETSU [6], Elleson and Freund [7] and 420
Singh [8] but less than those reported by Werner et al. [9]. Exclud- 421
ing the upper and lower 10% values gave minimum SECs that 422
were more similar to those previously reported. However, when 423
the data are compared, the results confirm the large range in 424
SECs for cold stores where for chilled and frozen stores the least 425
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
8 J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx
Fig. 4. Predictions from model (scenarios 1, 2 and 3) and range in energy consumption (dashed lines) superimposed onto survey data for chilled stores.
efficient stores used 4–5 times more energy than the most effi-426
cient stores. This indicates that considerable energy savings are427
possible.428
Much of this variation can be explained by shape factor of the429
store, usage and to a lesser extent ambient conditions surrounding430
the store. When using a mathematical model to understand differ-431
ences in energy use a large proportion of the survey data could be432
explained by these factors. When survey data was outside of the433
model predictions the store sizes tended to be small. This would434
indicate that use of these stores varied from the scenarios mod-435
elled or that a factor of their design affected their efficiency. As436
most of these stores used more energy than the model predicted it437
would seem likely that high usage and inefficiency contributed to438
the high energy usage reported.439
The performance of all stores (chilled, frozen and mixed)440
was statistically different. However, there was more relationship441
between the performance of frozen and mixed stores than there 442
was between chilled and frozen or chilled and mixed stores. The 443
energy used by chilled stores was less than frozen or mixed stores 444
at volumes below 22,000 m3 but was higher above this value. This 445
might indicate that large frozen stores tend to be long term stores 446
with less usage and that larger chilled stores have high usage (e.g. 447
large regional distribution centres where food is moved in and out 448
of the store many times per day). 449
It would be expected that larger stores would be more efficient 450
and have a lower SEC than smaller stores. This was found to be 451
the case by Werner et al. [9]. In this work the indications were 452
that this was only the case for frozen and mixed stores. For chilled 453
stores the relationship between volume and store size was linear. 454
The model demonstrated that SEC did vary with store volume but 455
that it was most apparent at low store volumes and that at store 456
volumes above 10,000 m3 for chillers and 20,000 m3 for freezers 457
Fig. 5. Predictions from model (scenarios 1, 2 and 3) and range in energy consumption (dashed lines) superimposed onto survey data for frozen stores.
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx 9
Fig. 6. Modelled impact of store total surface area (for scenarios 1, 2 and 3 at range of ambient temperatures) on annual energy consumption compared to survey data forchilled stores.
Fig. 7. Modelled impact of store total surface area (for scenarios 1, 2 and 3 at range of ambient temperatures) on annual energy consumption compared to survey data forfrozen stores.
the rate of change in the SEC was less than 0.5 kWh m−3 year−1 per458
10,000 m3 increase in store volume.459
The analysis demonstrated a surprising lack of relationships460
between the factors recorded (apart from volume) and annual461
energy consumption. There was for example no relationship for462
any store types with temperature of the store even though the463
range in temperatures recorded were relatively wide ranging (13 C464
for chilled and 5 C for frozen) and there was an extensive data465
set. In other instances the lack of any relationship may have been466
due to the restricted data sets available. It would therefore be467
useful to collect further data on the factors that were indicated468
to be important by the regression analysis and the mathematical469
model.470
5. Conclusions 471
Survey data demonstrated differences between chilled, frozen 472
and mixed usage cold stores. Store volume was the dominant factor 473
that was related to energy used by the cold stores. The impact of 474
cold store construction or usage had little impact on improving the 475
relationship between store volume and energy consumption. This 476
may have been due to a range of factors influencing energy con- 477
sumption which themselves had a high correlation with volume. 478
The mathematical model provided a better understanding of the 479
variations in cold store energy consumption and helped to explain 480
how usage, store shape factor and ambient conditions surrounding 481
the store contributed to the range in efficiencies in the survey data. 482
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
10 J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx
The model was shown to be a useful tool to estimate energy use483
of a cold store and provided a mechanism to generate metrics that484
can be used to assess efficiency of a cold store.485
Uncited referenceQ2486
[15].
Acknowledgements 487
The authors would like to thank EACI (Executive Agency for 488
Competitiveness and Innovation) for funding this work and in 489
particular the project officer, Christophe Coudun for his help in 490
managing the project. 491
Appendix 1. Information collected in survey 492
Survey page heading Units
Basic information:Total electricity usage for the system in the year reported?Does the electricity energy figure of the system submitted include?1. Compressor2. Lights3. Fans4. Pumps5. Fork/lift charging6. Blast freezing7. Floor heatingIf figure supplied includes blast freezing what is energy useEXCLUDING blast freezing?What is the total volume of the room(s) supplied by the system?What was the throughput in the year reported?What is the main function of the room(s)?What is stored in the room(s)?
What is the chilled set point temperature of room(s)?What is the frozen set point temperature of room(s)?Do you plans to invest in energy saving equipment?
kWh
Yes/No/Don’t knowYes/No/Don’t knowYes/No/Don’t knowYes/No/Don’t knowYes/No/Don’t knowYes/No/Don’t knowYes/No/Don’t knowkWh
m3
TonneChilled/frozen/mixed storage/blast freezing/loadingMixed foods/Meat/Fish/Fruit/Vegetables/Dairy/Cereal products
CC
Yes/No/Don’t know
Refrigeration system:Primary refrigerantRefrigerant quantity/chargeAmount of refrigerant added to primary system in the year reportedSecondary refrigerantRefrigerant quantity/chargeAmount of refrigerant added to primary system in the year reported
Don’t know/R22/CO2/Ammonia/OtherkgkgDon’t know/R22/CO2/Ammonia/Otherkgkg
Food stored:Average intake temperature for chilled productsAverage intake temperature for frozen productsDoes the room have controlled atmosphere?Does the room have humidity control?How is the food stored in the area?
How much food can be stored in the storage areaHow many pallets/containers can be stored in the storage areaWhat is the number of pallets/containers INTAKE in the year reportedWhat is the number of pallets/containers RELEASE in the year reportedWhat is the average size and weight of one pallet/container
CC (for mixed system fill both)Yes/No/Don’t knowYes/No/Don’t knowDon’t know/Pallets/Bins/Dolavs or containers/Placed on shelvesKg
NumberNumber
NumberWidth/Height/Depth all in m/kg
Facility:How many separate rooms does the system supply?What is the total floor area supplied by the system?How much of the floor area is used for:• Chilled storage• Frozen storage• Blast freezing storageHow many doors (total) are there on the room(s)?How many times on average will each door be opened per day?Do the doors have any protection?Is product automatically or manually loading into the room?Where are the room (s) positioned?What is the age of insulation?
What is the thickness of the:• Wall insulation• Ceiling insulation• Thickness of the floor
Numberm2
m2
m2
m2
Number
NumberDon’t know/No protection/Strip curtain/Air curtain or Airlock/Automatic doorsDon’t know/Manual (hand or fork lift)/Automatic (robot crane)Don’t know/Inside a building/OutsideDon’t know/ < 5 years/5–10 years/10–20 years/ > 20 years
mmmmmm
493
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx 11
Survey page heading Units
Refrigeration equipment:Type of refrigeration cycle?Type of refrigeration system?
Type of compressors?Do you have economised compressors?What is the compressor control system?How are compressors controlled?
Type of condensers?Defrost type?Do you use any heat from the refrigeration plant?• If yes, what forWhat is the year of installation of the system?
Don’t know/Single stage/Multi stage/Cascade/Absorption cycle/AircycleDon’t know/Dry evaporator with thermostaticvalve/Flooded-pumped/Flooded-natural circulationDon’t know/Reciprocating/Screw or Scroll/Rolling piston/CentrifugalYes/No/Don’t knowDon’t know/VSD/Unloading/OtherDon’t know/Suction pressure/Room air temperature/OtherDon’t know/Air cooled/Evaporative/Water cooled/Cooling tower/OtherDon’t know/Hot gas/Electric/Passive/Other
Yes/No/Don’t knowWater heating/Floor heating/Other heatingYear
494
Appendix 2. Cold store model495
Nomenclature496
A surface area (m2)497
e efficacy of the lamps (lm W−1)498
COP coefficient of performance of the compressor499
E effectiveness of door protection or blockage500
F density factor501
g acceleration due to gravity (9.81 m s−2)502
h heat transfer coefficient (W m−2 K−1)503
k thermal conductivity (W m−1 K−1)504
l latent heat of fusion for water (J kg−1)505
L height of cold store door (m)506
LF luminous flux (lm m−2)507
m mass flow rate (kg s−1)508
n stage coefficient509
N number510
M weight loss from product and packaging (kg day−1)511
P electrical power512
q heat flow (W)513
r respiration514
t duration515
T temperature (C)516
U overall heat transfer coefficient (W m−2 K)517
x fractional vaporisation of refrigerant in evaporator on518
expansion from liquid to saturation at discharge519
X concentration of water in air520
Greek521
empirical constant for different refrigerants522
thickness (m)523
density (kg m−3)524
efficiency525
Subscripts526
ad air through door527
c condensing528
comp compressor529
d door530
do door opening531
ds door seals532
do24 door openings per 24 h533
de defrost534
e evaporating535
f floor536
fl fork lifts537
fu fusion538
i inside539
l lights 540
m motor 541
o outside 542
ot other 543
pe personnel 544
pr product 545
T total 546
v vapour 547
w wall 548
Model 549
The model was steady state, therefore all heat loads were aver- 550
aged over one day. The shape of the cold store was a rectangular box. 551
There was only 1 door and the cold store was otherwise fully sealed. 552
The cold store had enough thermal mass such that door openings 553
did not change the temperature in the cold store. The temperature 554
of the ambient air outside the cold store was not changed by the 555
door openings. There was only one layer of insulation on the walls, 556
roof and floor. Any metal cladding was ignored as the resistance 557
to heat transfer from this was considered negligible. The luminous 558
flux from the lights was divided by the area of the floor and walls 559
to give a uniform luminance. The thermal mass of the trucks was 560
ignored. Therefore if they move from a warm environment into 561
the store, they do not give up this heat to the store. Energy from 562
fork lift trucks did not include charging the batteries. Any product 563
which changed temperature when loaded into the store did not 564
have a latent load (e.g. freezing and thawing) only a sensible load. 565
Respiration was included for all vegetable and fruit product above 566
0 C. 567
Data was input via a spread sheet. The inputs included; 568
• Information about each wall (including ceiling and floor) of the 569
cold store, e.g. face area, whether it was in the sun, outside ambi- 570
ent or internal and the type and thickness of the insulation. 571
• The size of the door, its opening schedule, whether it was pro- 572
tected (e.g. by strip or curtains), amount of traffic through the 573
door and the outside conditions. 574
• The refrigeration system, refrigerant, type of condenser, con- 575
denser ambient, efficiency of compressor and number of stages. 576
• Heat loads inside the store from forklifts, lights, personnel, prod- 577
uct, defrosts and evaporator fans. 578
• Electrical loads from lights, defrosts, evaporator fans and con- 579
denser fans. 580
Heat loads 581
The total heat load, qT, on the cold store was given by Eq. (1): 582
qT = qw + qdo + qde + ql + qfl + qpe + qpr + qm + qot + qf (1) 583
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
12 J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx
The heat load through the cold store wall was calculated using Eq.584
(2):585
qw = U · Aw · (T0 − Ti) (2)586
The overall heat transfer coefficient, U was calculated from Eq. (3).587
A surface heat transfer coefficient of 9.3 W m2 K−1 was used for hi588
and ho. If the weather was selected as windy, ho was increased to589
34 W m2 K1.590
1U
= 1hi
+ 1ho
+ w
kw(3)591
The heat load through the door opening, qd, was calculated using592
the sensible and latent heat exchange caused by mass flow of air593
during door opening and through the seals when the door was594
closed (Eq. (4)). The latent heat of fusion, lfu = 0 when the evapo-595
rating temperature was >0 C.596
qdo = (mdo + mds) · [cp(To − Ti) + (Xo − Xi) · (lfu + lv)] (4)597
The mass flow through an open door was calculated using the Gos-Q3598
ney and Olama model (Eq. (5)) [21]. An effectiveness value was599
used to reduce the infiltration for door protection devices and traf-600
fic obstructing the opening as detailed by Chen et al. [19]. The mass601
flow through the seals was a function of the condition and length602
of the door seal.603
mdo = (1 − E) · 0.221 · Adi
(1 − o
i
)0.5(g · L)0.5 · F (5)604
The density factor was calculated according to Eq. (6).605
F =(
2
1 +( i
o
)0.333
)1.5
(6)606
The heat load due to people was calculated from ASHRAE [17,18]607
(Eq. (7)).608
q = 273 − 6 · Ti (7)609
The heat load from forklifts trucks was calculated from Eq. (8). The610
model provided values for small, medium or large trucks, electri-611
cally or internal combustion powered.612
qfl = Nfl · Pfl · tfl (8)613
The product load was calculated based on flow of product into the614
store and the sensible heat it added or removed (Eq. (9))615
qpr = m.c.(Tp − Ti) + qr (9)616
The heat load of the condenser and evaporator fan motors, qm was617
given in Eq. (10). Where the electric motor were mounted outside618
of the cold store, m = 1.619
qm = NmP
m(10)620
The heat load from the defrost was given by Eq. (11):621
qde =(
1de
− 1)
·(
mad · (Xo − Xi) · l · t · Ndo24(M.l)24 · 3600
)(11)622
Electrical power 623
The total electrical power was the sum of all the electrical loads 624
(Eq. (12)) 625
ET = Ecomp + Econd + Eevap + Edef + Ef + Eo (12) 626
An electrical energy of the compressor, Ecomp, was derived from the 627
total heat load (Eq. (1)) using a calculated coefficient of performance 628
(COP) (Eq. (13)). The COP of the refrigeration system was calculated 629
using the formula given in Cleland et al. [20] (Eq. (14)) 630
Ecomp = qT .COP (13) 631
COP = (Tc − Te)(273 + Te)(1 − ˛.x)ncomp
(14) 632
The electrical power of the condenser and evaporator fan motors, 633
Em was the same as the heat load given by (Eq. (10)). 634
For electric defrosts, the electrical power of the defrost heater 635
was given by Eq. (15). If the defrost was hot gas or natural Ede = 0. 636
Ede = 1de
[mad.(Xo − Xi).l.t.Ndo24] + (mwat.l)(24.3600)
(15) 637
The electrical power of the lamps El was given in Eq. (16). 638
El = Lf.Af + Aw
e(16) 639
The total calculated heat load was presented plus individual heat 640
loads from transmission, infiltration (door opening), defrost, lights, 641
fork lift trucks, personnel, product, evaporator fans and other heat 642
loads. The total electrical energy was presented plus the individual 643
electrical loads from the refrigeration compressor, defrosts, con- 644
denser and evaporator fans, lights and floor heating. 645
References 646
[1] J. Guilpart, Froid et alimentation: Sécurité, sûreté ou procédé, in: Conférence 647
Centenaire du froid, Paris, 2008. 648
[2] Renewable Energy Technology Roadmap 20% by 2020, EREC (European Renew- 649
able Energy Council), 2008 (November). 650
[3] S. Mudgal, B. Tinetti, J. Bain, R. Cervantes, A. de Prado Trigo, Preparatory Study 651
For Eco-design Requirements for EuPs Lot 1. Task 2: Economic and Market 652
Analysis, 2011. 653
[4] J.E. Duiven, P. Binard, Refrigerated storage: new developments, Bulletin of the 654
IIR 2002-2 (2002). 655
[5] J. Bosma, Inventory study of the energy conservation potential in cold stor- 656
age insulations in the Netherlands, in: Proc. 19th International Congress of 657
Refrigeration, vol. II, 1995, pp. 382–391. 658
[6] ETSU, Energy Consumption Guide 37: Cold storage sector, Energy Efficiency 659
office, Department of the Environment, UK, 1994. 660
[7] J.S. Elleson, S.W. Freund, Benchmarking the energy performance of industrial 661
refrigeration systems, in: Interim Draft Report, Industrial Refrigeration Con- 662
sortium, College of Engineering, University of Wisconsin-Madison, US, 2004. 663
[8] R.P. Singh, Benchmarking study of the refrigerated warehousing sector in 664
California, in: Public Interest Energy Research Program, California Energy Com- 665
mission, 2006. 666
[9] S.R. Werner, F. Vaino, I. Merts, D.J. Cleland, Energy use in the New Zealand cold 667
storage industry, in: IIR-IRHACE Conference, Auckland, 2006. 668
[10] A. Carlsson-Kanyama, M. Faist, Energy Use in the Food Sector: A data survey, 669
Environmental Strategies Research Group/Swiss Federal Institute of Technol- 670
ogy, 2000. 671
[11] BELF, Bundesministerium für Ernährung, Landwirtsch, Energie und 672
Ernährungswirtschaft. Teil I: Bericht über die Energiesituation im 673
Ernährungsgewerbe, Teil II Energieverbrauch für die Herstellung aus- 674
gewählter Lebensmittel, Reihe A: Angewandte Wissenschaft, Heft 285, 675
Landwirtschaftsverlag Gm, Münster-Hiltrup, 1983. 676
[12] J. Swian, Reducing the Energy Cost of Potato Storage (Final Report), Britich Potato 677
Council, Report No. 2010/14. 678
[13] J.A. Evans, A.J. Gigiel, Reducing the energy consumption in cold stores, in: The 679
22nd IIR International Congress of Refrigeration, Beijing, 2007. 680
[14] J.A. Evans, A.J. Gigiel, Reducing energy consumption in cold storage rooms, in: 681
IIR ICCC, Cambridge, 2010. 682
[15] Weatherbase, Canty and Associates LLC, 2013, Available from 683
http://www.weatherbase.com/. (accessed 16.07.13). 684
[16] J.A. Evans, E.C. Hammond, A.J. Gigiel, L. Reinhold, K. Fikiin, C. Zilio, Improving 685
the energy performance of cold stores, in: 2nd IIR Conference on Sustainability 686
and the Cold Chain, Paris, 2013. 687
Please cite this article in press as: J.A. Evans, et al., Specific energy consumption values for various refrigerated food cold stores, EnergyBuildings (2013), http://dx.doi.org/10.1016/j.enbuild.2013.11.075
ARTICLE IN PRESSG ModelENB 4676 1–13
J.A. Evans et al. / Energy and Buildings xxx (2013) xxx–xxx 13
[17] A. Foster, C. Zilio, M. Corradi, L. Reinholdt, J. Evans, Freely available cold store688
energy models, in: 2nd IIR Conference on Sustainability and the Cold Chain,689
Paris, 2013.690
[18] ASHRAE, ASHRAE Handbook Fundamentals, 1997.691
[19] P. Chen, D.J. Cleland, S.J. Lovatt, M.R. Bassett, An empirical model692
for predicting air infiltration into refrigerated stores through doors,
Int. J. Refrig. 25 (September (6)) (2002) 799–812, http://dx.doi.org/ 693
10.1016/S0140-7007(01)00089-5. 694
[20] A.C. Cleland, Polynomial curve-fits for refrigerant thermodynamic properties: 695
extension to include R134a, Rev. Int. Froid. 17 (4) (1994) 245–249. 696
[21] W.B. Gosney, H.A.L. Olama, Heat and enthalpy gains through cold room door- 697
ways, Proc. Inst. Refrig. 72 (1975) 31–41. 698
top related